Tuesday, June 23, 2015

Science Matters for the non-scientist

Science Matters for the non-scientist - part 1

I am indebted to a person who prefers to remain anonymous who transcribed shiurim of mine. This virtually unedited version will serve some purpose I hope.

Science Matters for Non-Scientists
Rabbi Dovid Gottlieb
Science matters – matters can be either a noun or a verb – for the non-scientist; that means everything will be explained, hopefully in clear enough terms that if you have no science background you’ll be able to understand.
One of the reasons that it’s important to look at science is because of the enormous prestige that science enjoys in our society, the common western culture. That prestige has effect far beyond what I think people appreciate, and it was graphically revealed by a series of experiments initiated by Stanley Milgram of Yale University in the ‘60s, although he interpreted it differently and he had a different purpose for it from the moral that I will draw. I think you will be able to see that the moral that I draw is at least as appropriate as his.
The experiment works like this: there are three people who are involved in the experiment. The one who’s running it, who’s called the experimenter, and there’s someone who’s supposed to be learning and there’s a teacher.
Now, this is the way it looks – it looks one way, the reality is another way – but this is the way it looks: two people present themselves as volunteering to participate in this experiment. Psychological experiments on campus are very common and you’d pay students, in those days it was two dollars an hour, and they did whatever you wanted – sorted shapes or listened to sounds. And they are told that this is an experiment in learning, and in particular it’s using negative feedback. In other words, someone is supposed to learn something and he’ll be tested on it, and when he fails he’ll be punished, to see the effect of punishment on the ability to learn.
Here’s how it’s done. What he’s learning is word pairs. The learner is in one room, the teacher is in a different room. They can’t see one another, though there’s a wall between the two rooms and, as you’ll see, it’s possible to hear through the walls, but you can’t see. The teacher speaks into a microphone, the learner has earphones, and the teacher recites a list of word pairs. And then he goes back through the list one by one and for each first member of the pair he then gives four options to choose from and the learner chooses one option.
If he gets it right he goes on to the next entry. If he gets it wrong, the learner is punished by an electric shock. The teacher administers the electric shock. That is to say, the teacher has a control which has various settings for voltage and presses a button that administers the shock. It starts with fifteen volts and then each time the learner makes a mistake, the voltage goes up fifteen points. If he gets it right he just goes on to the next pair.
Before the experiment starts, the teacher is given a shock from the instrument, a fifteen volt shock, which is trivial, just so he feels that it’s real.
When the two volunteers present themselves, the experimenter hands each one a slip of paper and they are told, on one slip of paper is written the word “teacher” and on the other slip of paper is written the word “learner”; it’s presented to them as if the choice of the piece of paper is arbitrary. Then the learner is taken off into another room. Now, the learner makes mistakes, and for each mistake he makes he gets another shock and the voltage goes up.
At a certain point, the learner starts to moan and groan in pain and at a certain point he starts to yell and scream in pain and then he starts banging on the wall saying, “Stop this! Stop this!” And as you could imagine, the teacher begins to show signs of discomfort and hesitation, at which point the experimenter encourages the teacher. The first statement of encouragement is “Please continue,” and if he still shows hesitation, he’s told, “The experiment requires that you continue.” And if he still doesn’t do it, the third statement is, “It is absolutely essential that you continue.” And if he still hesitates, he’s told, “You have no other choice, but you must go on.” If after all four of those encouragements the teacher still refuses to go on, the experiment is stopped.
Now, that’s the way it looks. The truth is, there are no electric shocks and nothing’s being taught and the guy in the other room who’s designated as the learner is an actor who’s trained to act as if he were getting shocks and the yells and screams are recorded and they’re being played back from a recording. And the two slips of paper both have the word “teacher” written on them, so it guaranteed that this schnook over here is the teacher; the other guy who’s the actor says he got the piece of paper with the word “learner” on it, but of course it isn’t true.
The only subject in this experiment is the teacher. How far will he go? How far will he go in administering electric shocks to someone who, for all he can tell, is really suffering from these shocks? The top voltage is 450 volts, which is certainly fatal, and if the teacher will administer three 450 volt shocks in a row then the experiment is also stopped; in other words his contribution is stopped.
The first time he did it, he did it with Yale students. He asked random Yale students and his colleagues in the psychology department, “What do you think the result of the experiment will be? How many will go to the maximum voltage?” Both his colleagues and other Yale students said it would be one or two percent. As a matter of fact, 65% were willing to go to the maximum voltage.
Of course they had to be encouraged and they did show signs of discomfort and at one point they would ask, “Maybe we should stop?” and so forth and so on, but 65% went to the maximum voltage. And this experiment was repeated in other universities and other places around the world and the results were consistent. The ones who were prepared to go all the way to fatal voltages were 61-66 percent, regardless of time or place.
Furthermore, as another observer of these experiments pointed out, even those, the 30% or so, who refused to go to the fatal shocks, none of them said the experiment itself should be stopped. They didn’t protest, they didn’t threaten to go into the newspapers to have it shut down. They didn’t want to do it, but they didn’t threaten to shut it down. Nor did any of them initiate going to check on the condition of the supposed learner who’s the victim of the shocks to see if he’s okay. So, although they stopped short of administering fatal shocks, they didn’t threaten the experiment itself; they just didn’t want to go that far themselves.
Now, I’m giving out some of the other variations that were taken, but what was the point of the experiment and what explanation or what interpretation did Milgram and his colleagues put on it? It was testing obedience to authority. This was right after the Eichmann trials in Jerusalem and they wanted to see how people respond to authority; could a person be so influenced by authority to give up his otherwise moral convictions – none of these people we assume would randomly inflict pain on other people and certainly not to the point of a fatal shock, but here, in this circumstance they gave up their own convictions. With misgivings and with worries, yes, but they did it, they carried it out to the end. 65% carried it out to the end.
Now, Milgram says he was analyzing response to authority. And one could apply it to the Nazis and one could apply it to various authority structures. That’s what he says. But between you and me, he only tried one source of authority and that is the scientists. He didn’t try the governor of the state. He didn’t try the local subway designer, he didn’t try a mathematician, he didn’t try chess players, he didn’t try football players; he only tried scientists. To draw conclusions concerning the nature of response to authority in general from an experiment that only tested scientific authority, I think is a big jump.
And I am not alone: In his book Irrational Exuberance, Yale Finance Professor Robert Shiller argues that other factors might be partially able to explain the Mailgram Experiments: “[People] have learned that when experts tell them something is all right, it probably is, even if it does not seem so. (In fact, it is worth noting that in this case the experimenter was indeed correct: it was all right to continue giving the "shocks"—even though most of the subjects did not suspect the reason.)”

I don’t know how they would’ve responded in a democratic country to elected authorities, where a person could be president of the United States for 8 years and after that he’s just your local Tom, Dick and Harry. Of course he gets paid 2 ½ million dollars for every lecture that he gives, but other than that he’s just another person. You see Bill Clinton on the street, you’re not going to bow down to him, and if he offers an opinion, you could tell him to his face that’s he wrong. He was president, but he’s nobody anymore.
I think that this indicates narrowly the enormous prestige that science has in our culture, to the extent that when a scientist pressures a person to do something which otherwise would violate his moral norms, they are prepared to do it.
And by the way, the experiment was done with men and with women and the results were the same. And indeed, in one case where they tried to vary the experiment and use a dog as the subject rather than a human being, there all of those who refused to participate to the end were men; the women participated all the way to the end. So if anybody has any sexist presuppositions as to whether women will or men will, that’s an interesting variation on the experiment.
I think that this makes an examination of science and its status, its reliability, and its place in our general thinking an extremely important matter to investigate.
Some of the things that I want to discuss with you. I want to take a look at the reliability of scientific findings, though, as we will see, that word is prejudicial, the assertions of science, what science says it has discovered – that word is also prejudicial as we’ll see – to see how reliable they are, science and scientists. We will talk about the age of the earth and we’ll talk about evolution. We’ll talk about the status of science, given the Jewish picture of how the world really runs. Since God is creating the world at every moment, the world is running on a foundation utterly different from that which science thinks it is describing and discovering. If so we should ask what position science has, what relevance it has, what status it has.
I will spend some time talking about scientific evidence for the soul – scientific evidence for the soul, as opposed to against the soul, which people would think would be the natural finding. I think there’s considerable evidence – scientific, as understood philosophically – and we’ll talk about that as well.
Now, I’m not trained in science, so if I make an assertion about science itself, I will always be quoting a recognized authority on science. But I am trained in philosophy, and I have published research in philosophy, particularly philosophy of mathematics. And I have studied philosophy of science, and when it’s a matter of the philosophical implications of the scientific material then I consider myself competent to report my own judgments, though of course I will be trading on the judgments of other philosophers as well; I will not go out on a limb and take a position that’s unique and against what the rest of the philosophical community would say. That’s an introduction to where we’re going.
Question: In an experiment, the expert is the scientist, so obviously they’re going to listen to the expert. If you took a different situation where a rabbi was the expert, of course somebody would listen to him. Maybe it’s not just because it’s the scientist, but just because he’s an expert in that field.
Answer: Maybe. But I’m asking what the experiment shows. The experiment shows that scientists have this kind of enormous authority; it doesn’t show that rabbis have the authority. It seems to me, one ought to do another experiment with rabbis to see whether it’ll be taken that way or not. That’s why I said, a democratic elected official, who has authority in the democratic country, might not have the same kind of response from people who he tells what to do; I don’t know.
“Maybes” are too cheap. I want to know, what does the evidence show? He wanted it to be an indication how people response to authority structures. I’m pointing out that the only authority structure he tested is scientific authority, that’s all we know about. Maybe the other is true, maybe it isn’t; let’s leave it at a maybe.
When I was teaching at Johns Hopkins, there was a professor in the physics department who discovered a particle. Okay, the more energy you have, the more particles you find. There may be an infinity of them; it’s not so terribly amazing. But he was considered to be a top, top physicist. So the student newspaper interviewed him for his views on religion. Well, they printed the article, somebody in the philosophy department cut it out and put it on the bulletin board with the caption, “see how many mistakes you can find in this.”
He’s trained in physics; he’s not trained in religions. So why would anybody be interested in his views on religion? Because if he’s a physicist, he’s super smart, super rational and therefore his views are super right, even if they aren’t about physics.
And when I tell it to you that way, it sounds stupid, as it should, but the editors of the Hopkins student newspaper didn’t see it that way. They thought it would be of relevance and important for people to know what this physicist’s views on religion are. But of course it’s just nonsense.
I think that kind of attitude, which people do have, shows that science has enormous authority, carries authority in the society; this experiment certainly supports that contention, and that’s why I think it’s important to examine exactly what science can and cannot do.
What I want to start with is scientific information. You could divide science broadly into a method and results. The scientific method is a matter of gigantic controversy. There’re some philosophers who say there is no method to science, and if you invent a method you’re putting it under constraints that it doesn’t deserve. You’re going to shackle people and they won’t be able to find new theories. There’s no method at all to science. And there are others who think that there is a method and they describe it in various ways. It’s a gigantic, difficult subject. Why the method should be successful is a difficult subject. I’m not talking about that, I’m not going to go into that.
What I want to talk about is information that science presents as discovered, found, scientific findings – scientifically verified information. And I want to ask: how reliable is that information? What you get in the normal science textbook. If you go to a science museum, a science museum presents you with exhibits which acquaint you with how the world goes, what it is, how it’s made, what it’s mechanism are.
I remember, when we were living in Baltimore, the science museum had an exhibit about how the continents got their shapes and their relative positions. They had people from history, each person was a cartoon figure with a cartoon bubble with his view – obviously making fun of them – and finally you have the exhibit of plate tectonics. Plate tectonics is the current description of how the continents got their position and their shape. Of course the idea was, look at all these people in the past with great names, who made such foolish remarks about it, and now we know the truth.
If I had been designing that experiment, I would have had another panel. I would have dated it “2050” and I would’ve put a question mark: this is what we’re saying today, but what will be in 2050? They don’t have that in science exhibits in science museums. They show you the truth. Ay, it may not turn out that way? There’s no uncertainty, there’s no worry that maybe the foundation might not be correct, it’s just a presentation of information.
Now, I want to look at this information that science presents, what you get in the New York Times science supplement: “We report to you that the laboratories found this and the astronomers found that”, and see what the reliability of this information is. I think it is of variable reliability. Not all of it is the same: some of it’s more reliable; some of it’s less reliable.
I would divide this information into four categories on a continuum from most to least reliable. I will describe the four categories and show you why the first is more reliable than the second, second that the third, third than the fourth, and then we will draw a moral from this distinction of different level of reliability.
The best case is repeatable, observable phenomena. Things that you can observe with your own senses over and over and over again. And where science says, when you observe X you can expect to observe Y. These things are either things that happen by themselves spontaneously: the sun comes up, the sun goes down, the stars come out, the seasons change, the plants grow in the spring and they die in the fall, if you’re in the northern climate, animals reproduce and grow and develop and die, human beings are born and grow and die. These are things which you can observe because they happen over and over again, they’re repeatable and observable. Or they’re things you can make happen. You could hard boil as many eggs as you want. Heat up the water, put the raw egg in and watch it happen. You can break glass as many time as you want. You can bend metal. You can plant seeds in different environments and check how they grow, because you can control it.
Repeatable, observable phenomena. And science says, when you have this particular observed feature then you can expect that particular observed feature because they go together. That’s where science is at its strongest. But even there it could make mistakes. Even when it’s its strongest it could make mistakes. Water boils at a hundred degrees centigrade. But, gosh, I was mountain climbing, and when I tried to hard boil my egg, I made a fire and I put the pot with the water on it and brought the water to a boil, I put it in for 5 minutes and I took it out and it was soft boiled. It wasn’t hard boiled, as I am used to at home.
Answer? Not on a mountain top. On a mountain top the water doesn’t boil at a hundred degrees centigrade, it boils at a lower temperature. So it doesn’t really boil at a hundred degrees centigrade, it boils at a hundred centigrade at sea level. Okay, let’s put that in.
You know, I tried it again and this time the egg came out hard boiled after three minutes! That’s because you used salt water. With impurities, the water won’t boil at a hundred degrees centigrade; it’ll boil at a higher temperature. Oh, it has to be pure water; okay, I’ll put that in the rule: pure water at sea level.
You know, I tried it the other day and it didn’t work. Well then, tell me about your water. Actually, I had the water on a potter’s wheel and it was spinning around. No, no, the water has to be still, it can’t be spinning around; it won’t work if it’s spinning around. Oh, so it’s still, pure, at sea level. Now, how many other conditions do we need? We wouldn’t have known that it has to be still unless someone had tried it on the potter’s wheel and discovered that it doesn’t work. How many other conditions are being used that we don’t know about? Do we ever have a guarantee that we figured out all the conditions?
Here’s another experiment. Take a container full of water in the shape of a cylinder, flat bottom, and in the center of the bottom there’s a hole with a stopper. A small, circular hole right in the center. Fill the container with water – it’s open on the top – let it sit for a while, a few days, and then remove the stopper from the bottom, so the water will drain out the bottom. And as it’s draining, the water will start to rotate in a counter-clockwise motion; that is in fact what it will do.
“Now, I tried this,” someone will tell you. “I tried it in New York, in London, in Paris, in Munich, in Moscow, in Jerusalem, in Teheran, in Beijing, in Tokyo, and the Philippines; I really worked on this. I tried it in a dozen places separated by thousands of miles, east and west, and it always did the same thing. So I now believe, I have very good reason to think, that that’s what it does: when you have a cylindrical container and you let the water sit for a while so it still, it doesn’t have any motion of its own, and you have a hole in the center, you take out the stopper, as it drains it’s going to start moving in a counter-clockwise motion. And then we try it in Johannesburg, and there it goes clockwise.
Gosh, did you think that the equator is a borderline for fundamental forces of nature, that below the equator things are going to be different from above the equator? I wouldn’t have guessed that. It just turns out that that’s the way it is. You would not have guessed, having tried it in all the initial places, that Johannesburg will be different; it just turns out that it is. You have to recalculate what it is that’s doing it and why it happens. Do we ever know that we have all the relevant conditions? How would we know that? We would have to try testing all possible variations.
So, even when we’re dealing with repeatable, observable phenomena, where we’re at our most confident, we can’t be sure that we have tried all possibilities.
They discovered that a bird that’s hatched in the nest, and grows up normally the way birds do, will build a nest next year, even though it never saw a nest built. And we do not believe that the chirping of the parent bird is giving them instructions for next year’s construction project. But a bird that’s hatched in a laboratory and fed and released to the wild will not build a nest. Something about being hatched in nest causes the gelling of a particular part of the brain which contains the capacity to build a nest next year.
Nobody would have guessed that if we hadn’t observed it. We would have assumed that nest building is hardwired into the brain of the bird, that’s why it doesn’t have to see a nest build in order to be able to do, and since it’s hardwired it doesn’t matter where it develops. But it turns out not to be true.
Now, how do we ever know that we have all the conditions? We never know that we have all the conditions. So, even with repeatable, observable phenomena, where science is at its strongest, there’s still the possibility of mistake.
The next step is what’s called interpolation. Interpolation means this: let’s say I am testing something and I test it a number of time under slightly different conditions to see how it works when I change the conditions. So, if I think about the conditions I tested, I know what happens, and I have reason, we’ll take for granted, to assume that when I try it again, under the same conditions, it will happen again.
What happens when I try a new experiment when I’ve changed the conditions, what do I know about that? One could be very hard-headed here and say, “Nothing, you haven’t tried it yet, so you don’t know anything about it.” But let’s take a particular case. Here’s what I’m doing. I’m taking a cube of sugar and dissolving it in a glass of water, and I’m charting how long it takes for the sugar cube to dissolve in the water. And I’m trying it at different temperatures of the water: 10 degrees centigrade, 30 degrees centigrade, 50 degrees centigrade, 70 degrees centigrade, 90 degrees centigrade. Of course, you know what will happen – the warmer the water, the faster it will dissolve. But, I’m talking about exactly how fast it dissolves. I tried it a bunch of times at 10 degrees, and it takes, in this glass of water and this size lump, 2 minutes. And I tried it at 30 degrees and it takes a minute and two thirds. At 50, a minute and one third. At 70 degrees, a minute and at 90 degrees two thirds of a minute. Something like that. The rate it dissolves gets faster and faster as I go up the scaled.
What will happen if I try it at 20 degrees centigrade? I haven’t tried it at 20 degrees; I tried 10, 30, 50, 70 and 90, I didn’t try 20. So, as I said, a hard-headed person could say, “I don’t know, I haven’t tried it yet, wait till we try it.” But, no scientist will do that; what he’ll say is this: make a graph. This is the temperature, this is the rate of dissolving and the dots look like this. They go up because it dissolves faster at each time. 10, 30, 50, 70, 90. Now, draw a line connecting the dots. You know, little kids do that, connect the dots. Now, ask for 20, for which there’s no dot. 20, where does it intersect the line? And that’ll tell you how fast it will dissolve at 20.
In other words, I’ll use 10, 30, 50, 70 and 90 to predict 20 on the grounds that it changed regularly and therefore when it goes through the 20 point, it’ll be right where the line is.
That’s called interpolation. It’s called interpolation because I have the 10 and the 30 on both sides, this is in between; that’s what interpolation means – put something in between other things.
Some imaginative people have pointed out, you have there five dots. There are a lot of different ways to draw a line to connect the five dots. Lots and lots of ways. Who says it has to be a straight line? You could draw wavy lines, like a sine curve, or a co-sine curve, which is the opposite, or any kind of scribble to join them. And of course, if you draw a different line connecting the dots, the prediction for 20 degrees is going to be different. It will depend upon where 20 degrees hits the line. If it’s a sine curve it’ll be very high up and you’ll predict that it’ll be dissolving very fast, if it’s a co sine curve it’ll be very far down; another one will have a different shape. How did you pick the straight line to connect the dots? And the answer of course will be, what you all know, that it’s the simplest line connecting the dots.
Yes, it is the simplest line connecting the dots, and that’s definitely the right answer. It’s just that, as of now, anyway, there’s no agreed upon definition of simplicity. What exactly does simple mean? How should simple be defined? It’s a subject that people have thought about and discussed and debated and there’s no satisfactory definition of what simplicity means.
Number two, why should the simpler line be preferred? Why? I have read people who write, it’s just aesthetically more pleasing; I like it better, it’s more satisfying. Isn’t that a really extraordinary thing to say? I’m predicting what’s going to happen in the real world. I’m going to drop this cube of sugar into 20 degrees and before I do so I’m going to predict this is how long it’s going to take to dissolve. Why? Because I like that line better; it’s more aesthetically pleasing to me; it’s more satisfying; I get warm fuzzies when I look at that line, and that’s why when I drop the sugar into the water, that’s how fast it’s going to dissolve. Isn’t that a little anthropomorphic to think that the world cares what’s aesthetically pleasing to me?
So, just to sum up and then I’ll take your questions. This is where we stand: the judgments of simplicity are usually agreed upon. And when we test them, the judgments of simplicity usually turn out to be correct. We just don’t have any definition of simplicity and we don’t have any explanation of why simplicity should be relevant. But those are merely philosophical concerns; they don’t affect the scientific practice in this matter.
Therefore I am going to describe this category, category number two, interpolation, as very secure – not as secure as repeatable, observable phenomena, because there you actually saw what you’re predicting – but it’s very secure; it’s just slightly less secure than the repeatable, observable, phenomena.
Question: If you can test the simple line, then why challenge it? If you take the simple line and connect the dots and see where 20 is and then test that, if it’s correct then obviously the simple line is the best thing to use.
Answer: Let’s see. You had five dots and we observed that there’s actually an infinity of different ways of drawing lines to connect the five dots. Now you say: let’s test the simplest line and we test it. But now, there’s also an infinity of ways of connecting the 6 dots, and every one of those was equally tested by your test of the 6th dot because every one of them is on your 6th dot.
Let’s say it in detail. We have 10, 30, 50, 70, 90, and we’re testing 20. Now we say: before we test, let’s make a prediction. What predictions could we make? A straight line, or the wavy curve, which give different answers for 20. You’re thinking of lots and lots of different lines which give different answers for 20. And I say, “Okay, here we are. We have the straight line answer for 20 and all these other weird curves and they give different answers for 20. Let’s test it and see what happens.” Correct. So we test it and it turns out that the straight line is the correct one. The straight line is correct against all of the competitors which gave different answers for 20. What I want to point out to you is that when you add a 6th one, you could repeat the same problem. Consider all the different ways of joining the 6 dots, including the 20 dot. You may do it again, but in the meantime you have no evidence to predict which one you should pick. You only ruled out the crazy lines which differ at 20. There are lots and lots of crazy lines that don’t differ at 20; you didn’t test those.
Question: If you test it for the 5th one and then you test for the 6th one and you test it for the 7th one, and all three of those work, then you could figure that the 8th one, the 9th one, and the 10th one are also going to work.
Answer: But you understand, when I draw the crazy line, it fits the 6th, 7th, and 8th also; it fits all of them. So how do I know what it predicts is correct? It fits the straight line and all the other crazy lines that go through all eight. It fits all of them; so the question is, how do you pick which one that fit all eight. Any finite set of points can be drawn in an uncountable, infinity of ways. So whenever you do it, true you’ve ruled out some – that’s correct; you’ve ruled out all the crazy ones that differ at 20 – but there’s an infinity of crazy ones that agree at 20, and you haven’t ruled those out.
Question: If there’s a finite amount of dots which can be placed, that wouldn’t imply that there’s an infinite amount of ways to draw the dots, because you would eventually get every single possible dot.
Answer: It’s not a question of implying; it’s a mathematical fact that there’s an infinity of dots. What you could do, theoretically, is do so many experiments that it’s impossible to physically discriminate the inputs anymore; that could be done, but it would be a gigantic, gigantic expense. It’s never been done, and we make predictions confidently without doing that. The goal here is to explain how we make our predictions and what they’re based on. In that case, you would reduce case two to case one; that’s true. But that would be something that no one has ever done.
Question: So you’re saying, why would you use a straight line, only for the simplicity of it. If that’s true, why would you draw on any other formula ever? They govern certain different laws that you’re trying to figure out one way or another…
Answer: There are times when I try and I get the opposite result. I try the straight line and my observations ruin the straight line; they violate the straight line. For example, if I chart radioactive decay, it’s not going to work out with a straight line; it’s going to be a decreasing curve. Or acceleration; if you're taking about accelerating speed or motion, then it’s not going to work out that. Then I’m going to have to talk about the simplest type of curve that fits the point for this type of phenomena, where it’s not a linear phenomenon.
Question: In the thing you’re dealing with it’s pretty clear it’s two variables.
Answer: you’re quite right. If you have an algebraic representation of the graph, then you’ve already decided which curve you want. We’re trying to induce that from the data. That’s why I said you could have a linear line or a sine or a co-sine. They are algebraically very different from one another. To decide an algebraic representation, is to already have chosen a particular way of joining the dots. My question is, how do you choose the way of joining the dots?
The number of variables isn’t the issue here; in radioactive decay, you only have one variable. You have the number of radioactive atoms as a percentage of the totality of the substance. There’s no other variables there. And it still has a non-linear curve. It’s not the number of variables that counts; it’s the type of phenomenon that it is. We know from experiment that the different types of phenomena get different types of curves and the way it curves. All that is after I’ve decided to analyze the data one way rather than another.
The key to all of these decisions ultimately is simplicity, and as I said, I’m not challenging simplicity as the right criterion – it is the right criterion – I’m just pointing out that there are two subsequent philosophical questions that don’t have answers, but I’m inclined to say that’s not so important, even though I’m a philosopher. We do use it. There’s very regular agreement as to which is the simpler interpretation. Well, maybe I should stop and expand that a little bit.
I should say that when we talk about simplicity, I’m talking about the simplicity of joining dots on a piece of paper. When you deal with explaining how the world works, it’s not so easy. You want to know what makes a typhoon, or what causes the economy to go up and down and so forth and so on. Somebody proposes a theory and somebody proposes another theory, and people say, in general you should choose the simplest theory. There it’s much worse. What makes a theory more simple than another theory? So here people talk about Occam’s razor. Occam was a medieval philosopher and he said – of course they’re saying in translation, it’s not clear what he said, but it’s usually reported – don’t multiply entities beyond necessity. When you say you’re going to explain how it happens, you say, “Well, there are atoms and then there are molecules, and then there are forces between the atoms and the molecules, and then there are particles…” That’s a lot of stuff. Don’t put any stuff in unless you really need. If somebody can get by and explain it with less stuff, it’s a better explanation, because it’s simpler.
But then some people will tell you, it’s not the number of things you put it; it’s the number of assumptions you make. How many forces are there? Forces aren’t things; forces aren’t stuff; forces are how things react to one another. Now, let’s say you have one theory witch 17 things and 3 forces and another theory that has 5 things and 9 forces. Which one is the simpler one? Do forces and things count equally? Is one more important than the other? Not a clue. And some people count the number of fundamental concepts and others count how familiar the concepts are and on and on……..
Again, in concrete cases, people are pretty confident and there’s pretty widespread agreement which count is the simpler theory. But if you ask why, and you try to concoct a definition of what counts as the simpler theory, it’s very difficult to come by.
Also it may reflect your commitments. Stephen Hawking writes that when they discovered the universe is expanding and the natural thought was, “Gosh, if it’s getting bigger, than in the past it was smaller” – that sounds right – “and if you go further back in the past, it gets smaller and smaller. There’s got to be a limit to this; it can’t get smaller than zero. So, it’s got to have a beginning. If it’s really been getting bigger and bigger and in the past it was smaller and smaller, it’s got to have a beginning.”
Well, as a matter of fact, in the 1930s they avoided that conclusion. They found a way to describe a universe that’s expanding in some sense or other and never goes back to zero. In fact, let me say it a little more carefully, according to their theory, things are getting further and further away from one another, but the whole is always staying the same size. How do you do that? You say it’s infinitely big. Space is infinitely big; there are no limits. And it’s filled with an infinity of stuff scattered around. It’s true, things are rushing away from one another; that’s what the red shift shows. Everything is getting further and further away. But there’s plenty of room; you’re not going to run out of room, because there’s infinite room. Ay, our world doesn’t seem to be getting thinner and thinner? That’s because everywhere new stuff is coming into existence out of nothing, ex nihilo, in order to fill the spaces. That was the Steady State Theory – the universe is infinitely big, everything is rushing away from everything else, but new stuff is coming in in the middle, and don’t worry that you don’t see it because in a cubic meter of space, in a hundred years or so, one atom will come into existence. So it’s not as if you’re going to have new petunias in your backyard over night; the rate of appearance of matter is very, very slow; we’ll never be able to observe it.
Why did they opt for that? Infinitely big world and new stuff coming into existence out of nowhere? Because, as Stephen Hawking writes, we didn’t want a beginning, because a beginning smacks of supernatural intervention; those are his words. And therefore we opted for a theory that avoids a beginning. Which theory is simpler? An infinite space, which you don’t observe, and new stuff coming into existence everywhere in the universe, which you don’t observe; aren’t you multiplying things here? You’re multiplying new, creative episodes all over the universe, which you don’t observe, and an infinity of space, which you don’t observe, to avoid what? To avoid a beginning. Is a beginning so terribly complicated, that it deserves to be avoided by postulating infinite space and new stuff coming into existence all over the universe? I don’t know whether that’s an obvious judgment of what’s more simple and what’s less simple. That sounds like, as he says, we don’t want a beginning.
Question: That’s just because you raise up questions. It’s not simply the beginning; it’s something like where is the beginning and what started it. It’s because they don’t have the answers for that.
Answer: If you’ll ask the parallel questions about the infinite universe, like where is the stuff coming from that’s coming into existence ex nihilo throughout the universe? I think if you’re going to ask about the origin at the zero point, you ought to be asking about the origin of all this stuff that just appears from nothing throughout the universe. And you could ask what sustains the existence of the universe as a whole. It’s not obvious that the questions are not parallel. I’m just pointing out that these are delicate judgments. I think it’s very remarkable that when people make these judgments, they agree; even though you can’t spell out what your criteria is; you can’t spell out what your definition is, people do agree. Since they do agree, I say it’s reasonable to rely upon. That’s why I say it’s very reliable; not as reliable as repeatable, observable phenomena, but very reliable, and I’m not going to issue any critique of this second category, just to point out that it’s got question marks about its basis.
Question: Assuming that the whole thing you’re saying is true, that the universe was started from zero and it’s continually getting bigger, would that denote the fact that as it gets farther away or as it get bigger things come into existence faster?
Answer: You mean according to Steady State Theory. Steady State Theory was canned in 1965 because they discovered the background radiation which indicated the universe did have a start from a very small state.
Question: That doesn’t get rid of the Steady State Theory.
Answer: But then you don’t need it. The idea of things coming into existence was only to plug a hole. If you say that space is infinite and things are rushing away from one another, and you stop there, then things ought to be getting thinner and thinner through time. The conglomerations of matter ought to be getting thinner and thinner through time. There ought to be bigger and bigger distances between items through time. We don’t see that; we don’t observe that. How do I counter the fact that space is not thinning out? Answer: new stuff is coming in throughout. We didn’t observe that, but we’re just using that to plug the hole of “how come things aren’t thinner and thinner”. Once you give up the idea that space is infinite and things have been expanding and running away from forever, and you say the universe had a beginning, and a beginning a finite amount of time, I don’t need to plug the hole with that. It was a speculation put in just to plug a hole.
Question: Is it that far-fetched to think that things come into existence out of nowhere?
Answer: We never see that; in fact, according to contemporary science you have the law of conservation of mass energy – when you start an experiment and end an experiment you have the exact same quantity of mass energy at the beginning and the end; that never changes.
Question: I’m talking about the bare things that come in and out of existence.
Answer: I don’t know what you mean by existence. If a thing came into existence, then in that box there would be a net gain in mass energy; but it never happens. Here we’re talking about a net gain. That’s never been observed. Every experiment that we have performed and every observation we’ve made, mass energy remained constant. If you will discover a change in mass energy of an experiment you’ll win the Nobel Prize, for sure. Your name will be written in the science books for at least 100 years. And by the way, that wasn’t so solid; 150 years ago they though mass never changes and energy never changes. Both of those turned out to be wrong; it’s the sum total of mass energy that doesn’t change; that’s because of Einstein’s relativity.

To create a theory in which new mass energy is coming into existence throughout the universe all the time, when in every experiment and every observation we make there’s never any change in mass energy, is a gigantic step. The idea is just to plug a hole to avoid a beginning.